AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Cross-Resolution Distribution Matching for Diffusion Distillation

Chen Feiyang, Hongpeng Pan, Haonan Xu, Xinyu Duan, Zhefeng Wang, Yang Yang

ECCV 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add cross-resolution distribution matching (RM) to a cascaded 512->1024 student (ablation)RMD on SDXL · HPS v2.1 promptsHPS: 29.57 -> 31.99Improved
add RM at low-resolution target (no upsampling) (ablation)RMD on SDXL · HPS v2.1 prompts (512px target)HPS: 22.64 -> 30.28Improved
add RM to cascaded video generation (ablation)RMD on Wan2.1-14B · VBenchVBench total: 81.11 -> 82.51Improved
add cascaded upsampling (UP) on top of RM (ablation)RMD on SDXL · HPS v2.1 promptsHPS: 30.28 -> 31.99Improved
use 2+2 step allocation instead of 3+1 (ablation)RMD on PixArt-alpha · HPS v2.1 promptsHPS: 24.79 -> 32.13Improved
use 1+3 instead of 2+2 allocation (ablation)RMD on PixArt-alpha · HPS v2.1 promptsHPS/speedup: 32.12 -> 32.13 HPS; 1.25x vs 1.68x speedupNo clear effect
mix predicted noise and Gaussian noise (alpha 0.2) in re-injection (ablation)RMD on PixArt-alpha · HPS v2.1 promptsgeneration fidelityImproved
go from two to three resolution stages (256->512->1024) (ablation)RMD on SDXL · HPS v2.1 promptsHPS vs speedup: 31.99 -> 31.59 HPS; 1.56x -> 1.73x speedupMixed
go to four resolution stages (128->256->512->1024) (ablation)RMD on SDXL · HPS v2.1 promptsHPS: 31.99 -> 29.48Hurt
replace 4-step single-resolution TDM with 2+2 cascaded RMDSDXL distillation · HPS v2.1 promptsHPS average / speedup: 31.93 -> 31.99 HPS; 20.0x -> 33.4xImproved
RMD vs 6-step distilled video baselinesWan2.1-14B · VBenchTotal score: 80.48 -> 82.51Improved
RMD vs the undistilled teacher on videoWan2.1-14B · VBenchTotal score: 83.75 -> 82.51Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy